[mythtv-users] Firewire_tester giving weird segfaults and failing to allocate bandwidth

[Gabe Rubin]

I have been ongoing struggles with what used to be a very stable
firewire connection.  After I thought I thought I solved everything,
and made some succesful recordings, the connection once again became
unstable.

When I use firewire_tester, I got the following output:
[mythtv at localhost ~]$ /usr/local/bin/firewire_tester -p -n 0 -r 5
Action: Test P2P connection 5 times, node 0, channel 0
P2P: Testing...Success, 407 packets received
P2P: Testing...Success, 238 packets received
P2P: Testing...Success, 431 packets received
P2P: Testing...Success, 373 packets received
P2P: Testing...libiec61883 warning: Failed to allocate bandwidth.
Failed

Once I get that failure, I can't get the firewire connection working
again unless I reconnect the firewire cable.  I have tried resetting
the box.

I am also getting a ton of these in my mythbackend log:
                        eno: No such device (19)
2010-02-10 17:47:11.551 LFireDev(005094000005B6C8), Error: raw1394_loop_iterate
                        eno: No such device (19)
2010-02-10 17:47:11.552 LFireDev(005094000005B6C8), Error: raw1394_loop_iterate
                        eno: No such device (19)
2010-02-10 17:47:11.552 LFireDev(005094000005B6C8), Error: raw1394_loop_iterate
                        eno: No such device (19)

Somewhere along the way, dmesg or /var/log/messages was reporting a
segfault for firewire_tester.

Wondering if my attempt to fix the problem with a new box is causing a
new set of issues.